Dr. Martina Seiffert

© dkfz.de

Dr. Martina Seiffert
Team Leader
Contact: → Contactform
Tel.: 06221 42-4586

Immune microenvironment in B-cell lymphoma

Figure 1: Schematic presentation of major components of the tumor microenvironment and their interactions with the malignant B cells in CLL.
© dkfz.de

Survival and proliferation of B cell lymphomas is regulated by extrinsic factors provided by the microenvironment (Hanna et al. 2017). In Chronic Lymphocytic Leukemia (CLL) for example, peripheral blood B cells are cell cycle arrested, whereas in the bone marrow and lymph nodes, proliferation of CLL cells is induced by the local niche (Seiffert et al., 2010 and Schulz et al., 2011). But the interaction between cancer cells and non-malignant cells is a two way street: the tumor shapes its microenvironment to be more favorable including the severe skewing of myeloid and T cells towards leukemia-supportive and immunosuppressive phenotypes. We aim to unravel the mechanisms behind this communication, asking questions like: What mediates the signals between the cells? Which cells in the microenvironment are providing the pro-tumor signals? Specifically, how is the immune system involved?

Novel drugs that target interactions of malignant cells with non-malignant stromal cells or block immune regulatory proteins show encouraging results in clinical trials with cancer patients. As CLL and other B-cell lymphomas are modulating their mircoenvironment to be protective by providing supportive clues and inhibiting effective immune cell activation, targeting the cross-talk of malignant cells and their niche might result in mobilization of tumor cells, induction of apoptosis and re-activation of anti-tumor immune activity to eradicate residual disease. In our work we aim to identify and test new targets to modify the tumor microenvironment and enable tumor elimination.

Role of exosomes in modulating the tumor microenvironment

Figure 2: A. Electron microscopy image of CLL-derived EVs. B. Uptake of fluorescently-labelled exosomes of the CLL cell line MEC-1 (green) by macrophages (DAPI staining of nucleus = blue). C. Upregulation of PD-L1 expression measured by flow cytometry in monocytes that were treated with CLL-derived exosomes. D. Proposed mechanism of how exosomes induce an inflammatory and immunesuppressive microenvironment that supports CLL development. See Haderk et al. 2017
© dkfz.de

Extracellular vesicles (EVs) are important mediators of intercellular communication and involved in many physiological and pathological processes. Protein and RNA molecules contained in the membrane-enclosed nanoparticles can be transferred to a target cell via fusion or endocytosis. Within the target cell the transferred material is active and can cause cellular changes. EVs released by tumor cells represent a novel mechanism which allows the cells to communicate and thereby alter their microenvironment.

We established a standardized protocol based on serial centrifugations to isolate exosomes from blood plasma of CLL patients and healthy donors as well as culture supernatant of CLL cell lines. Characterizing CLL-derived EVs by electron microscopy and Western Blot revealed vesicles with the typical cup-shape and size of 30-350 nm, which were positive for exosome markers RAB5a and HSP70 (Haderk et al., 2013). Exosomes carry an exosome-specific protein and RNA profile with which the tumor shapes its microenvironment. Analysing this further, we could show that noncoding Y RNA hY4 is highly abundant within CLL-derived exosomes. This non-coding RNA induces strong, TLR7-dependent phenotypical changes, like cytokine release and PD-L1 expression, in the main target cells for CLL-exosomes: monocytes and macrophages (Figure 1), thereby creating an inflammatory and tumor-supportive milieu (Haderk et al. 2017).

Ongoing work is currently focusing on deciphering the role of tumor-derived exosomes in the development and progression of CLL and other B-cell lymphoma.

Immunosuppressive mechanisms of myeloid cells

Figure 3: A. Adoptive transfer of malignant splenocytes from Eµ-TCL1 mice in syngeneic receipients leads to development of CLL-like disease. B. Patrolling monocytes quantified by flow cytometry based on expression of Ly6C and CD43 accumulate in the spleen of mice after transplantation (Tx) of TCL1 leukemia. C. Depletion of myeloid cells in TCL1 tumor-bearing mice using clodronate liposomes led to decreased spleen weight, a major site for malignant B-cell accumulation, compared to control animals receiving PBS liposomes. See Hanna et al., 2016.
© dkfz.de

To study the full complexity of microenvironmental networks in CLL in vivo, we use the transgenic Eµ-TCL1 (TCL1) mouse model for CLL (Bichi et al.,2002). In these mice, development of CLL is associated with a systemic inflammatory cytokine milieu and alterations within the distribution and activity of myeloid cells (Hanna et al., 2016), which confirms our previous observations in human CLL (Seiffert et al.,2010; Schulz et al., 2011). Monocytes and macrophages are skewed toward pro-tumorigenic phenotypes, including the release of tumor-supportive cytokines and the expression of immunosuppressive molecules such as programmed cell death 1 ligand 1 (PD-L1). Depletion of monocytes and macrophages in these mice slows disease development and repairs T cell defects, highlighting the oncogenic properties of myeloid cells in CLL.

 

Our future studies aim to decipher the pathogenic role of myeloid cells, which show a tremendous plasticity. Tumors can induce a variety of tumor-supportive and immunosuppressive cell phenotypes like tumor-associated macrophages (TAMs) and myeloid-derived suppressor cells (MDSCs). Single cell analysis will highlight the role the different cell types play.

Mechanisms of T-cell exhaustion

Figure 4: A. Leukemia develops faster in Rag2-/- mice that lack T and B cells compared to wild-type control mice, which leads to their earlier death. B. CLL development in mice is associated with an accumulation of oligoclonal CD8+ effector T cells, based on their TCR sequence, that show heterogeneous PD-1 expression. C. Single-cell RNA-sequencing of CD8+ T cells from leukemic TCL1 mice revealed 8 distinct clusters of cells based on unsupervised clustering of differentially expressed genes using Seurat package (Satija Lab), which could be plotted over “pseudotime” as suggested by Monocle (Trapnell Lab). A. See Hanna et al., 2018; B+C. Manuscripts in preparation
© dkfz.de

-cell responses play a major role in controlling tumor growth. Without T cells, CLL develops much faster, as seen when Eµ-TCL1 leukemia is adoptively transferred into Rag2-/- mice, which lack mature B and T cells. Tumor-bearing Rag2-/- mice die earlier then tumor-bearing wild-type controls. Pin pointing the T-cell subset responsible, CD8+ effector T cells seem to control CLL development (Hanna et al., 2018).

In the long run, these tumor-specific effector T cells fail to fully eradicate the malignant cells. They rather show signs of activation-induced exhaustion including enhanced expression of inhibitory receptors like PD-1, KLRG-1, 2B4, and LAG3, impaired immune synapse formation, and decreased cytotoxicity both in mouse models (McClanahan et al., 2016; Hanna et al., 2018) and CLL patients (Ramsay et al., 2008; Riches et al., 2013).

We now aim to decipher mechanisms of T-cell exhaustion. We have already identified markers allowing sub-grouping of functionally distinct T-cell subsets for detailed characterization. Additionally, we are investigating different microenvironmental clues such as IL-10 and their role in T-cell exhaustion. Using single cell RNA-sequencing, we further obtain a higher resolution of cellular differences and a better understanding of the function of individual T cells in the context of the microenvironment of B-cell lymphoma.

Preclinical testing of immunotherapy

Figure 5: Treatment with anti-PD-L1 antibodies slows down leukemia development upon adoptive transfer of TCL1 leukemic cells in syngeneic mice. Reduced spleen weight, a major site for malignant B-cell accumulation, and a lower percentage of leukemic cells in the spleen, peripheral blood (PB), bone marrow (BM) and peritoneal cavity (PC) were observed. See McClanahan&Hanna et al., 2015
© dkfz.de

Given the role immune cells and mediators play in the microenvironment of B-cell lymphomas (see above sections), one of the focuses in our lab is to test and validate the potential of new therapeutic drugs that target these immune components.

Work from us and others showed that CLL and its associated alterations within the microenvironment are very well mirrored in the Eµ-TCL1 adoptive transfer (AT) model. Therefore, we use the TCL1 AT model to analyze the pathogenic relevance of cell subsets, candidate genes or pathways, identified in our studies looking at myeloid and T cells. One approach to test the importance of different cell types, is the specific depletion of cell subsets using antibodies or transgenic mouse lines.  Further, we test compounds that target identified gene candidates or signaling pathways, either as monotherapy or in combination with already approved lymphoma drugs such as ibrutinib, PI3Kd inhibitors, or immune checkpoint inhibitors.

Immune checkpoint blockade, which aims to re-activate “silenced” immune cells, is a new therapy approach for lymphomas. Using the TCL1 AT model, we showed that targeting the PD-1/PD-L1 immune checkpoint results in improved T-cell function and leukemia control in mice (McClanahan et al., 2015).

Mechanisms of resistance to targeted therapy and immune checkpoint blockade

Figure 6: A. Treatment of leukemic mice after adoptive transfer of malignant TCL1 cells with the Btk inhibitor ibrutinib (Ibr) leads to control of leukemia development, measured as the number of CD5+CD19+ cells in blood, for 3 to 4 weeks. Thereafter, mice develop drug resistance with a rapid expansion of leukemic cells. B. One week after treatment start, ibrutinib impairs proliferation of leukemic cells as measured by Ki-67 staining, which was lost 6 weeks after treatment start. Manuscript in preparation
© dkfz.de

Treatment with novel targeted drugs like the Bruton’s tyrosine kinase (Btk) inhibitor ibrutinib or the Bcl-2 antagonist venetoclax improved outcome for CLL patients considerably. But resistance to therapy is increasingly observed and remains an urgent clinical need. Immune checkpoint blockade with antibodies targeting PD-1 or PD-L1 were not effective in initial clinical trials with CLL patients, but the reasons for this unexpected failure of response are unclear.

We study mechanisms of therapy resistance in the TCL1 AT mouse model by comparing treatment-sensitive and -resistant leukemia cells, as well as the immune microenvironment in responsive and non-responsive mice. The obtained results will be compared to data from clinical trials with the goal to assess the potential and to improve the outcome of rationally designed combination therapies, including immune checkpoint inhibitors, as treatment strategy for CLL patients.

Selected Publications

to top